Astron. Astrophys. 346, 811-818 (1999)

2. Observations and data reduction

2.1. The Beppo-SAX mission in brief

The X-ray satellite Beppo-SAX (Boella et al. 1997a) is equipped with four co-aligned Narrow Field Instruments: the measurements that we report here were obtained by using the LECS (Low Energy Concentrator Spectrometer, Parmar et al. 1997), two of the three MECS (Medium Energy Concentrator Spectrometers, Boella et al. 1997b), and the PDS Phoswich Detector System (PDS, Frontera et al. 1997). The LECS and MECS have imaging capabilities and cover the energy bands 0.1-10 keV and 1.7-10 keV, respectively. The LECS and 2 MECS instruments provide in the overlapping band an effective area that is three times the area provided by a single LECS/MECS unit, with similar energy and spectral resolution. The two instruments together allow us to study best the Fe K complex at 6.7 keV, which is an important diagnostic of both the coronal abundance and the temperature structure in the coronal sources. Moreover, the LECS resolution, which is comparable to the resolution of CCD detectors at low energies, and its sensitivity at energies down to 0.1 keV, where typical coronal sources have the largest X-ray flux, render the LECS data especially reliable for the study of coronal plasmas. In fact, the spectrum of a coronal source in the region below the carbon edge (E0.3 keV) has few and relatively week lines and is characterized by a high signal-to-noise ratio. This relatively line-free region, as demonstrated by Favata et al. (1997a), allow us to constrain most accurately the global metallicity of the coronal plasma.

The operative energy range of the PDS is 15 to 300 keV, and this instrument can be used for both spectral and temporal studies.

2.2. The Beppo-SAX observations

The Beppo-SAX observations were obtained on 1997 Nov 2-4, and on 1997 Nov 11-12. The run Nov 11-12 was done to complete the planned observations that on Nov 4 were interrupted by a Gyro Scientific Mode (GSM) fallback during satellite orbit No. 7714. The effective exposure times are listed in Table 1. The significantly shorter-than-scheduled exposure time for the LECS instrument occured because, at the dates of observations, the LECS could be operated only during Earth dark time.

Table 1. Observation log of Beppo-SAX AR Lac observations on Nov 1997.

2.3. The data analysis

The data analysis was based on the linearized, cleaned event files obtained from the Beppo-SAX SDC on-line archive. Light curves and spectra were accumulated using the XSELECT Ver. 1.3 tools, using 8.5 and 4 arcmin extraction radii for the LECS and MECS, respectively, that include more than 90% of the fluxes. The resulting source counts are listed in Table 1. The background LECS and MECS spectra were extracted from the appropriate event files available from the SAX SDC public ftp site, containing the standard background observations, obtained by summing long pointings of sky regions with no detectable sources. The light curve analysis was performed using the XRONOS 5.1 package, while the spectral analysis was performed with the XSPEC 10.0 package (Arnaud 1996), implemented under FTOOLS 4.0. We used the response matrices released by the ASI SDC in September 1997. For spectral analysis, the LECS data have been analyzed only in the 0.1-4 keV range, due to still unresolved calibration problems at higher energies. We used addspec under FTOOLS to generate one total LECS and one total MECS spectrum by summing for each detector the two spectra observed during the two observing runs. In the following we will refer to these added spectra with the code (T). To look for short term variability of the coronal plasma, we also analyzed the two separate spectra acquired starting on Nov 2 and on Nov 11, respectively. It was not useful to further dissect the observations, mainly because the lack of LECS data during the second half of the first observing run.

The LECS and MECS spectra have been jointly fit after allowing for a rescaling of the LECS data to account for uncertainties in the intercalibration of the instruments. This rescaling factor value was determined as a free parameter of the fit. All spectra were re-binned in order to have at least 20 counts per energy bin. The background subtracted source mean count rates are given in Table 1.

© European Southern Observatory (ESO) 1999

Online publication: June 17, 1999
helpdesk.link@springer.de